The premise of this resesrch is that current smartphones power saving strategies can be improved towards saving more power and or gain more user satisfaction only if they start following “preventive” and or “user customized” power saving plans. This resesrch develops a number of preventive power saving strategies which help in saving the battery power without the need of using the power of the same battery for detecting abusers (detective strategies). Although, the focus of this work is on power-aware smartphone network wireless connections, other layers such as application, operating system and hardware layers of smartphones are of important interest to optimize our solutions across all layers. Smartphones can’t work as stand-alone devices. They have to connect to a wireless network to do something useful. Bending network availability smartphones use on-demand strategy to connect to a specific network. This comes with a great deal of power consumption cost in addition to the cost of using the network if any. How to optimize this on-demand networking strategy using “preventive” and or user “customized” power saving plans acros multiple layers and wireless interfaces is the research of this study. A set of “preventive” customized power-aware connecting strategies are offered. Some of them are evaluated experimentally.

The Hidden Markov Model (HMM) lies at the heart of the modern speech recognition systems as it provides a simple, effective and straight forward frame work to model the time varying acoustic features of the speech signals. The basic process of building HMM based speech recognition systems is a straight forward process. Nevertheless, the proper parameter estimation of such models requires large training data. Therefore, parameter tying techniques were developed to reduce the parameters of HMMs without affecting the overall system performance. This study proposes an Arabic phonetic decision tree necessary to build Tied State tri-phone HMMs. Experimental results show promising word correctness when compared with both data driven tri-phone models and phoneme based models. The maximum word correctness achieved by the proposed approach was 95.13%. Whereas it reached 78.03 and 58.45% using data driven tri-phones and phoneme based HMMs, respectively, when tested on the same benchmark database.

Hyperspectral Imagery (HSI) is widely used in the application domains such as agriculture, environment, forestry and geology for the identification and observations which demands the efficient classification accuracy. The supervised classification is a challenging task due to limited number of available training samples compared to large number of spectral bands. This phenomena reduces the classification accuracy. To overcome this problem, the dimensionality reduction preprocessing step is adopted. This process reduces the number of spectral bands which leads to decrease in computational complexity and enhancement in classification accuracy. In this study, AEPCA (Auto Encoder and Principle Component Analysis) method is proposed for dimensionality reduction of HSI. The performance of AEPCA is evaluated against AE (Autoencoder) and PCA (Principle Component Analysis) method. The dimensionally reduced components are classified using CNN (Convolutional Neural Network) based classifier. The proposed model of dimensionality reduction demonstrates superior classification accuracy due to effective combination of characteristics of AE and PCA. The noisy or corrupted pixels are recovered by AE Model and high dimensional image is represented by efficient fewer number of principle components by PCA is the potential advantage of AEPCA Model.